JPH0965492A - Loudness compensation device and hearing aid device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a hearing aid device excellent in operability by which the user himself easily adjust the compensation of a loudness characteristics. SOLUTION: A test signal generating means 25 provides an output of a voice test signal while increasing the voice level gradually. The user inputs a reply signal to a terminal 10 when the voice is heard. The similar operation is applied as to test signals for plural kinds of frequencies. A hearing power calculation means 24 obtains the loudness frequency characteristics of the user's hearing power based on a voice level when the reply signal is received. A loudness compensation means 5 conducts compensation based on the loudness frequency characteristics. In this case, the compensation is conducted by using a transversal filter 21 to made compensation with high accuracy.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a loudness compensator used for digital processing of audio signals of a hearing aid such as a hearing aid or a sound collector, and a hearing aid using the same.

[0002]

2. Description of the Related Art In conventional hearing aids (for example, hearing aids), analog processing is generally used when filtering or expanding audio output of an audio signal.

However, due to recent improvements in digital signal technology, digital signal technology is being adopted for these hearing aids.

As a conventional example in which the digital signal technology is adopted for a hearing aid, a digital hearing aid "Digitarian" described in Reference 1 (Ono, "Current status and future of digital hearing aids", Acoustical Society of Japan, Vol. 47, No. 10, 1991). There is. The characteristic of this digital hearing aid is that the audio signal is processed by an analog filter into three frequency bands (low range, middle range,
High frequency range) and the gain of each frequency band is calculated by an external computer.
This is because it is possible to compensate the loudness characteristic to each user to a certain extent by setting the parameter according to the setting.

Further, another conventional example is described in Reference 2 (Yoshizumi et al., "Development of multi-signal processing type digital hearing aid"; Proceedings of the Acoustical Society of Japan 1-7, 16, 1991 / 10-11). There is a prototype of a digital hearing aid in use. The features of this prototype are the suppression of shock noises and the formant enhancement for clarifying speech, which are problems with the current hearing aids. Further, the audio signal is digitally divided into three frequency bands (bass range, middle range and treble range), and the gain of each frequency band is externally set in the same manner as in the above-mentioned conventional example. It is possible to compensate for the loudness characteristic of.

[0006]

Among the above-mentioned conventional techniques, a general hearing aid for analog-processing a voice signal is not provided with a loudness characteristic compensation function for each user.

Further, in a hearing aid which has adopted a recent digital signal technique, a voice signal is transmitted in three frequency bands (bass range,
There is one that simply compensates for the loudness characteristics by dividing into the middle and high frequencies. However, the following points are not sufficient for the user in terms of performance and usability.

(1) A simple loudness characteristic compensating method for setting the gain of each voice component divided into three frequency bands (bass range, middle range and treble range) based on a user's hearing test result. Therefore, it is difficult to perform accurate loudness compensation for each user.

(2) The gain of each frequency band for loudness characteristic compensation is set by an external computer (for example, a personal computer). Therefore, in order to check this characteristic compensation and update the gain, it is troublesome to ask a specialist, a specialty store, or the like.

The loudness characteristic of the user changes gradually. Nevertheless, the user thought that the loudness compensation characteristic once set was suitable for himself forever, and used it as it became unsuitable without noticing that the loudness characteristic of his hearing changed. There was something that happened.

As described above, it has been difficult for the conventional hearing aid to provide the user with the optimum compensation of the loudness characteristic. Further, the user cannot easily adjust the compensation of the loudness characteristic according to the intended use, which is inconvenient.

An object of the present invention is to provide a hearing aid in which the loudness characteristic can be adjusted by the user's own hand.

An object of the present invention is to provide a hearing aid which is more convenient to use.

It is an object of the present invention to provide a loudness compensating device whose loudness characteristic can be easily adjusted.

[0015]

Means for Solving the Problems The present invention has been made to achieve the above-mentioned object, and the first aspect thereof is as follows.
Hearing checking means for measuring frequency characteristics of user's hearing, loudness characteristic compensating means for performing loudness characteristic compensating processing on a voice signal inputted from the outside based on the measurement result of the hearing checking means, and the loudness compensating processing. Output means for outputting the sound signal subjected to
There is provided a loudness compensating device having:

The loudness compensating means is a transformer.
A monkey filter, a compensation characteristic calculation means for obtaining a frequency characteristic for loudness compensation necessary to compensate the frequency characteristic of the hearing ability of the user obtained by the hearing check means, and a loudness compensation for the loudness compensation obtained by the compensation characteristic calculation means. Obtaining the tap coefficient value of the transversal filter for realizing the frequency characteristic with the transversal filter,
It is preferable that the transversal filter includes a coefficient setting unit.

The above-mentioned hearing check means is provided with frequency and / or
Alternatively, a test signal generating means configured to generate a plurality of types of test signals having different output levels, and outputting only one of the test signals at a time, and the test signal generating means is outputting the test signal, Response signal receiving means for receiving and receiving an input of a response signal from the user, and the loudness frequency of the hearing of the user based on the frequency and the output level of the test signal output at the time of capturing the response signal. And a hearing acuity calculating unit for obtaining a characteristic, wherein the output unit outputs the test signal output from the test signal generating unit and not subjected to the loudness characteristic compensation processing to the outside. It is preferable.

It is preferable that the test signal generating means is capable of continuously changing the output level of the test signal.

A voice processing unit for performing a non-linear voice processing is provided for a voice signal input from the outside, and the loudness compensating unit compensates the loudness characteristic for the signal processed by the voice processing unit. It is preferable that the treatment is performed.

The above-mentioned nonlinear voice processing includes vowel pitch expansion processing, vowel pitch shortening processing, consonant emphasis processing,
It is preferable to include at least one of the above.

At least one of a voice synthesizing means for synthesizing voice and a voice recording / reproducing means for recording a voice signal inputted from the outside is provided, and the loudness compensating means is the voice synthesizing means and / or the above. It is preferable that the loudness characteristic compensation processing is performed on the output signal of the audio recording / reproducing means.

At least one of a voice synthesizing means for synthesizing voice and a voice recording / reproducing means for recording a voice signal inputted from the outside is provided, and the voice processing means is the voice synthesizing means and / or the voice synthesizing means. It is preferable that the output signal of the audio recording / reproducing means is subjected to the nonlinear audio processing.

As a second aspect of the present invention, a hearing ability checking means for measuring a frequency characteristic of a user's hearing ability, and a loudness characteristic compensation for an audio signal inputted from the outside based on a measurement result of the hearing ability checking means. A loudness characteristic compensating means for performing a process, an output means for outputting the audio signal subjected to the loudness compensating process to the outside, and a control means having at least a hearing aid mode as an operation mode and a hearing check mode. And the control means is
In the hearing aid mode, the loudness characteristic compensating means is caused to perform the loudness characteristic compensating process on the audio signal inputted from the outside, and in the hearing check mode, the frequency characteristic of the hearing ability of the user is checked by the hearing checking means. There is provided a hearing aid characterized by being measured.

The hearing checking means outputs a plurality of types of test signals having different frequencies one by one while gradually increasing the output level thereof, and outputs a response signal from the user during the output of the test signal. Is received, and the frequency characteristic of the hearing ability of the user is preferably obtained based on the frequency of the test signal output at the time of receiving the response signal and the output level thereof.

[0025]

[Action]

(1) User loudness characteristic measurement The hearing check means measures the frequency characteristic of the user's hearing. That is, the test signal generating means has a plurality of basic frequencies (for example, 100 Hz, 200 Hz, 500 Hz, 1 kHz, 2 kHz,
Outputs a test signal of a certain frequency among (). In this case, the output level (amplitude level) is initially set to a sufficiently low level. Then, the output level is gradually increased.

When the user can hear the voice of the test signal, the user inputs a response signal indicating that the test signal has been heard.

The response signal capturing means receives and captures this response signal.

The test signal generating means measures the hearing ability of the user for each frequency by executing the above operation for the test signals of a plurality of frequencies.

The hearing ability calculating means finds the frequency characteristic of the hearing ability of the user based on the hearing ability measurement result for each of these frequencies (for example, 100 Hz, 200 Hz, 500 Hz, 1 kHz, 2 kHz, ...).

(2) Calculation of Loudness Compensation Characteristic The compensation characteristic calculation means determines the frequency characteristic of the hearing ability of the user based on the frequency characteristic of the hearing ability of the user obtained by the hearing ability calculation means, for example, in a normal hearing person in a voice band. The frequency characteristic for loudness compensation necessary to match the characteristic of is obtained.

(3) Realization of loudness compensation characteristic The coefficient setting means, based on the loudness compensation frequency characteristic obtained by the compensation characteristic calculating means, taps for realizing the loudness compensation frequency characteristic by the transversal filter. Calculate the coefficient value.

The coefficient setting means sets the obtained tap coefficient value as the tap coefficient value of the transversal filter.

The output means outputs the audio signal subjected to the loudness compensation processing to the outside.

With the above operation, the user can measure the loudness characteristic of his or her own hearing, and the loudness compensation characteristic is automatically set.

In order to apply the present invention to a hearing aid, the hearing aid is provided with a hearing aid mode and a hearing check mode. The loudness compensation characteristic is set in the hearing check mode.

It should be noted that, by performing the loudness characteristic compensation process after the processing by the voice processing means (for example, vowel pitch expansion processing, vowel pitch reduction processing, consonant emphasis processing), the presence or absence of the voice processing processing is determined. Regardless, loudness characteristic compensation can be performed. Similarly, the loudness characteristic compensation process is performed after the process by the voice synthesizing unit and / or the voice recording / reproducing unit.

[0037]

Embodiments of the present invention will be described with reference to the drawings.

First Embodiment The features of this embodiment are as follows.

(1) One hearing aid normally uses a mode for hearing aid (hereinafter referred to as "hearing assist mode") and a mode for checking the hearing ability of the user (hereinafter referred to as "hearing check mode"). -").

(2) When the user's hearing is confirmed in the hearing check mode, the user is automatically checked according to the result.
The optimum loudness compensation characteristic is set.

(3) As a means for compensating the optimum loudness characteristic, a means of transversal filter type having a plurality of taps is provided, and it is possible to accurately compensate the loudness characteristic.

(4) Non-linear voice processing (for example,
Even if the execution or non-execution of consonant enhancement and vowel expansion) is switched according to the application, the loudness compensation characteristic provided to the user is constant.

The overall structure of the hearing aid of the first embodiment is, as shown in FIG. 1, an input processing section 2, a memory section 3, a voice processing section 4, and a loudness characteristic compensation processing section 5. A hearing check unit 6, an output processing unit 7 for converting the processed digital voice data into an analog voice, a bus unit 8 connecting these signal processing units, each input / output terminal 1,
It is composed of 9 and 10.

In FIG. 1, the thick solid line bidirectional arrows are digital data transfer lines between each signal processing means and the bus 8. The thin solid arrow lines indicate the flow of audio signals. The dotted line shows the flow of processing in each signal processing unit.

The input processing unit 2 converts an analog voice signal input from a microphone or the like into digital voice data. In this embodiment, the input processing unit 2 includes a band limiting filter (LPF or BPF) 11 and a clamp unit 1.
2, AD for converting analog signals to digital signals
It is composed of C13.

The memory section 3 is for storing digital voice data. In this embodiment, the memory unit 3
Is composed of a memory interface 14 for exchanging data with the bus 8 and a memory 15 for storing data.

The voice processing unit 4 performs various voice processing (for example, consonant enhancement and vowel extension for easy listening). In this embodiment, in order to realize the vowel expanding function and the consonant emphasizing function, the voice processing unit 4 includes a vowel pitch detecting means 16, a vowel expanding means 17, a consonant detecting means 18, a consonant emphasizing means 19, a vowel and a consonant. The synthesizing means 20 of FIG. Various voice processing techniques necessary for realizing the voice processing unit 4 have been proposed in the past and are well known techniques (for example, Japanese Patent Laid-Open No. 5-241598).
Since the audio processing technology itself mentioned here is not related to the essence of the present invention, the description thereof is omitted here.

The hearing check unit 6 is for measuring the hearing characteristics of the user. In the present embodiment, the hearing check unit 6 is composed of a test signal generation means 25 for hearing measurement, a response signal acquisition means 24, and a hearing calculation means 26.

The test signal generation means 25 generates a test signal used for hearing check. The test signal generation means 25 uses a plurality of basic frequencies (for example, 100 Hz, 200 Hz, 500 Hz, 1 kH) as the test signal.
z, 2 kHz, ...) Signals can be generated. Further, the audio output level of the test signal is automatically increased. Further, in response to an instruction from the response signal fetching means 24, the frequency of the test signal being output at that time and the audio output level at that time can be stored. The test signal generating means 25 sets the frequency and the audio output level to 1
It is possible to store not only the set but also the types of the fundamental frequencies that can be generated.

The response signal fetching means 24 detects the operation of a switch provided separately and receives it as a response signal. When the response signal is input (when the switch is operated), the response signal fetching means 24 informs the test signal generating means 25 to that effect. The switch itself does not necessarily have to be provided in the hearing aid itself. In this embodiment, the switch is configured to be removable, and is connected to the hearing aid (input terminal 10) only when the hearing check is performed.

The hearing ability calculating means 26 obtains the hearing characteristic of the user based on the sound output level for each frequency stored by the test signal generating means 25 in response to the response signal. The hearing characteristic is defined as a relationship between the frequency of the test signal and the minimum audio output level at which the user can detect the test signal of the frequency.

The loudness characteristic compensation processing unit 5 performs the loudness characteristic compensation according to the hearing characteristic of the user, so that the sound finally output from the hearing aid to the outside has a frequency characteristic that is easy for the user to hear. belongs to. In this embodiment, the loudness characteristic compensation processing unit 5
Is composed of a transversal filter means 21, a compensation characteristic calculation means 22, and a tap coefficient setting means 23 for setting the tap coefficient of the transversal filter means 21.

The compensation characteristic calculation means 22 calculates a compensation characteristic to be performed by the loudness compensation processing section 5 according to the hearing characteristic obtained by the hearing check section 6. The compensation characteristic calculation means 22 determines the gain of loudness compensation for each frequency so that the gain for each frequency reaches a preset level.

The transversal filter means 21 is a transversal filter composed of multi-stage taps.
A coefficient can be set for each tap. By appropriately setting this coefficient, the transversal filter means 21 can perform desired loudness compensation.

The tap coefficient setting means 23 sets the above-mentioned tap coefficient. The tap coefficient setting means 23
Is to obtain the value of the tap coefficient required to realize the compensation characteristic obtained by the compensation characteristic calculating means 22 and set it in the transversal filter means 21.
Note that the transversal filter and the method of obtaining the tap coefficient for obtaining the target frequency characteristic are well-known techniques, and thus the description thereof is omitted here.

The output processing unit 7 includes a DAC 27 for converting digital voice data into analog voice and a band limiting LPF.
(Or BPF) 28 and output amplifier 29.

Next, the operation of the hearing aid will be described for each operation mode.

(1) Operation in Hearing Aid Mode In this hearing aid mode, the audio signal input from the input terminal 1 is subjected to audio processing and loudness compensation, and then
Output from the output terminal 9.

The analog voice signal input from the input terminal 1 is first subjected to the band limiting filter 11 in the input processing section 2.
The voice band component is extracted by. Subsequently, the clamp 12 superimposes the DC voltage on the analog audio signal. Further, it is converted into digital voice data by the ADC 13. After this, this digital voice data is
It is transferred to the memory unit 3 through the bus 8 and stored in the memory 15 through the memory interface 14.

The voice data stored in the memory 15 is subjected to various voice processing by the voice processing unit 4.
For example, the pitch detecting means 16 and the vowel expanding means 17 increase or decrease the vowels by increasing or decreasing the number of pitches without changing the pitch width of the voice vowel. As a result, the input voice is converted into so-called "slow words" or "quick words". Further, the consonant detecting means 18 and the consonant emphasizing means 19 enhance the consonant part of the voice, thereby improving the word clarity. The voice data processed in this way is again stored in the memory unit 3 via the bus 8.

The voice data processed by the voice processing unit 4 is guided from the memory unit 3 to the loudness characteristic compensation processing unit 5 via the bus 8. Then, after the loudness frequency characteristic is compensated by the transversal filter means 21, it is guided to the memory section 3 (or directly to the output processing section 7) via the bus 8.

The voice data guided to the output processing unit 7 and having the compensated loudness frequency characteristic is converted into an analog voice signal by the DAC 27. In addition, the band limiting filter 28 extracts a voice band. Then, the signal in the extracted voice band is output from the output terminal 9 via the output amplifier 29.

The signal output from the output terminal 9 is guided to an earphone (not shown) or the like and converted into a sound there.

The sound provided to the user in this manner is a good one with loudness characteristic compensation.
Particularly, in the present embodiment, since the transversal filter is used, the accuracy of loudness characteristic compensation is excellent.

The loudness characteristic compensation processing unit 5 compensates the loudness frequency characteristic by a non-linear voice processing process (conversion into "slow words" or "quick words" by expansion / shortening of voice vowels) by the voice processing unit 4. It is performed after the processing of improving the intelligibility of words by enhancing consonants). Therefore, the loudness frequency compensation characteristic provided by the loudness characteristic compensation processing unit 5 does not change and is kept constant regardless of whether or not the nonlinear voice processing is executed.

(2) Operation in Hearing Check Mode In this hearing check mode, the hearing check of the user is performed using the test signal generated by the test signal generating means 25 of the hearing check unit 6.

The test signal generating means 25 generates a test signal. The audio output level of this test signal is automatically gradually increased from the minimum level (virtually stepless (continuous))
It can be made bigger. The test signal generation means 25 outputs the test signal to the output processing unit 7 via the bus 8. The output processing unit 7 performs the same processing as the hearing aid mode on the test signal and outputs the test signal from the terminal 9.

The user listens to the test signal sound output from the output terminal 9 via the earphone or the like. Then, when a test signal sound that gradually increases is heard, the switch connected to the input terminal 10 is operated. Then, according to the operation, a predetermined response signal is input to the response signal input terminal 10.

The response signal fetching means 24 has an input terminal 1
When the input of the response signal from 0 is detected, the fact is notified to the test signal generation means 25 and the hearing ability calculation means 26.

Upon receiving the notification that the response signal is input, the test signal generating means 25 stops the generation of the test signal of the frequency output at that time and starts the generation / output of the test signal of the new frequency. . Further, the voice output level at the time when the response signal is input is stored in association with the frequency.

Similar processing operations are repeated for the test signal of the new frequency. The frequencies of the test signals and the audio output levels thus stored are output to the hearing calculation means 26 after the hearing check is performed on the test signals of all the frequencies.

The hearing ability calculation means 26 calculates the frequency characteristic of the hearing ability of the user based on these hearing ability data, and transfers this to the loudness characteristic compensation processing section 5 via the bus 8.

The compensation characteristic calculating means 22 in the loudness characteristic compensation processing section 5 calculates the characteristic for compensating for the received loudness frequency characteristic. Then, the obtained characteristics are transferred to the memory unit 3 or the tap coefficient setting means 23 via the bus 8.

The tap coefficient setting means 23 obtains the tap coefficient of the transversal filter means 21 required for compensation based on the derived compensation characteristic data. The transversal filter unit 21 transfers the obtained tap coefficient to the tap coefficient unit of the transversal filter unit 21 (or the tap coefficient memory area of the memory 15 of the memory unit 3) via the bus 8. Tap coefficient unit (or memory unit 3
The memory area for the tap coefficient of the memory 15) holds the tap coefficient.

As described above, in the hearing check mode, the user himself / herself can periodically measure the hearing characteristics in the hearing check mode. Then, the hearing aid can automatically compensate for the loudness characteristic according to the measurement result. Therefore, the optimum loudness characteristic can always be provided to the user when used in the hearing aid mode.

The algorithm for automatic loudness characteristic compensation will be described again with reference to FIG. In the figure, the solid arrow indicates the signal processing algorithm in the hearing aid mode. Also, the dotted line arrow is an algorithm in the hearing check mode.

One of the hearing aid mode and the hearing check mode is set (step 10).
1). Here, when the hearing aid mode (102) is selected, the processing on the left side drawn by the solid line in the figure is executed. On the other hand, when the hearing check mode is selected, the processing on the right side drawn by the dotted line in the figure is executed.

That is, as described above, when analog voice is input (step 103), the band of the input analog voice is limited (step 104), and then the analog voice is converted into digital voice data. (Step 105). Then, the converted digital voice data is stored in the memory 15 (step 10).
6).

After that, processing processing is performed according to the purpose of the digital voice data (step 107). Further, the loudness characteristic of the processed digital audio data is compensated by the transversal filter means 21 (step 108). The digital voice data subjected to the loudness characteristic compensation processing is output via the bus 8 to the output processing unit 7
(Step 109).

In the output processing section 7, conversion of digital voice data into an analog voice signal (step 110), band limitation of voice converted into analog (step 111),
The band-limited analog voice is output (step 112), and the voice is guided to the user (125). In this case, since the optimization of the loudness characteristics of the transversal filter means is optimized in advance by the hearing check mode, the user can listen to a good voice.

On the other hand, when the hearing check mode (113) is selected in the mode setting (step 101), the processing on the right side in the figure drawn by the dotted line is executed.

That is, the frequency of the test signal for hearing check is initialized (step 114) and the output level (volume) of the signal is set to the minimum value (step 115). The test signal thus initialized is transferred to the output unit via the bus 8 (step 109).

After that, as in the hearing aid mode, the processing of steps 110, 111, and 112 is performed, so that
Provide the test signal audio to the.

When the test signal sound is heard, the user (125) operates the switch to input the response signal. The response signal fetching means 24 fetches this (step 11).
6). The response fetching means 24, based on the response signal,
It is determined whether or not the test signal sound being output at that time is heard by the user (step 117). If there is no response signal, it is considered not heard and the process proceeds to step 118.
In step 118, the test signal generation means 25 increases the test signal level. After this, again, step 110,
By executing steps 111 and 112, the user is made to hear the test signal sound whose volume has been increased again.

If there is a response signal in step 117, it is considered that the signal is heard and the process proceeds to step 119.

In step 119, the test signal generating means 25 records the frequency of the test signal being output at that time and its signal level. Then, it is further determined whether or not the hearing check has been performed for all the fundamental frequencies (step 120). As a result, if the hearing check has not been completed for all the fundamental frequencies, the process proceeds to step 121, and the frequency of the test signal is changed. Then, the test signal level is set to the minimum value for the new frequency (step 115), and the same processing is repeated again.

In step 120, if the hearing check has been completed for all the test frequencies, the test signal generating means 25 proceeds to step 122 to stop the test signal generation.

After that, the hearing aid calculating means 26 determines in step 1
Based on the audio output level for each frequency recorded in 19 (this is the minimum signal level that the user can hear for the audio of the frequency), the hearing characteristic of the user is calculated. , And outputs this to the loudness compensation processing unit 5 (step 123).

The compensation characteristic calculation unit 22 of the loudness compensation processing unit 5 calculates the loudness compensation characteristic suitable for the user based on the hearing characteristic input from the hearing calculation unit 26 (step 124). Further, the tap coefficient calculating means 23 calculates and sets the tap coefficient necessary for realizing the loudness compensation characteristic by the transversal filter means 21 (step 125).

As a result, in the above-mentioned hearing aid mode, the transversal filter means 21 performs the loudness compensation based on the loudness compensation characteristic set in this way.

As described above, in the first embodiment, the hearing ability can be easily checked by the user's own hand. Also, the loudness compensation characteristic can be automatically updated based on the result. Therefore, the user can always obtain the optimum voice.

In this embodiment, the audio output level of the test signal was automatically increased. However, it is also possible to accept an instruction from the user and change the output level according to the instruction. Instead of continuously changing the audio output level, it is possible to output one of a plurality of preset level test signals for the test signal of each frequency. In this case, with respect to the test signal of a certain frequency, the output levels of the test signal to be output do not necessarily have to be output in the ascending order of the output level. If the presence or absence of the response signal can be confirmed for each output level, the output order is not particularly limited.

Second Embodiment FIG. 3 shows the configuration of a hearing aid which is a second embodiment of the present invention.

The hearing aid of the second embodiment has the following new features in addition to the features of the first embodiment (see FIG. 1).

(1) In addition to the hearing aid mode and the hearing check mode, a mode for outputting a message from the voice synthesizing means (hereinafter referred to as "message mode") is also provided.

(2) A voice synthesizing means is provided, and it is possible to perform a voice processing process and a loudness compensating process on a voice synthesized by this voice synthesizing means.
The synthetic speech is used, for example, to announce to the user the usage of the hearing aid and the currently used mode.

In order to realize the above characteristics, in this embodiment, as shown in FIG. 3, the same device configuration as that of the first embodiment (see FIG.
In addition to the above, the voice synthesis processing unit 40 is provided in the preceding stage of the voice processing unit 4 and the loudness compensation processing unit 5.

Since the voice synthesis processing section 40 is a well-known technique, its description is omitted here.

The operation will be described.

Since the operations in the hearing aid mode and the hearing check mode are the same as those in the first embodiment, only the operation in the message mode will be described here.

The voice synthesis processing section 40 transfers the synthesized digital voice data to the memory section 3 (or directly to the voice processing section 4) via the bus 8.

The voice processing unit 4 converts the synthesized voice read from the memory unit 3 (or directly sent from the voice synthesis processing unit 40) into "slow words" or "quick words", Processing such as "consonant emphasis" and "pitch conversion" is performed. Then, the synthesized voice data after the processing is transferred to the bus 8
To the memory unit 3 (or the loudness compensation processing unit 5) via the.

The loudness compensation processing unit 5 applies the transversal filter to the synthesized voice read from the memory unit 3 (or directly sent from the voice processing unit 4) as in the operation in the hearing aid mode. The means 21 compensates the loudness frequency characteristic of the user. Then, the compensated data is output to the memory unit 3 (or the output processing unit 7) via the bus 8.

The output processing section 7 converts the synthesized speech read from the memory section 3 (or directly sent from the loudness compensation processing section 5) into analog data, and outputs it from the output terminal 9.

By the above operation, the explanation message of the hearing aid assisted by the voice synthesis processing unit 40, the alarm voice message, etc. are converted into "slow words" or "quick words", It is possible to improve the processing of word clarity by consonant enhancement. In addition, it is possible to perform accurate loudness characteristic compensation by the transversal filter means including multi-stage taps.

The operation in the message mode will be described again with reference to FIG. FIG. 4 shows the flow of processing in the message mode together with the flow of processing in the hearing aid mode. Since the flow of processing in the hearing check mode is the same as that in the first embodiment, the description in FIG. 4 is omitted. The operation in the hearing aid mode is the first
Since it is the same as the embodiment (see FIG. 2), the description thereof will be omitted.

When the message mode (201) is selected in the mode setting (step 101), the processing flow drawn by the one-dot chain line in the figure is selected.

The voice synthesizing unit 40 synthesizes voices such as an explanation message of the hearing aid and an alarm voice message (step 202), and the voice processing means 4 responds to this synthesized voice. Conversion processing to "slow words" or "quick words", consonant emphasis processing, etc. (step 1
07).

The loudness compensation processing section 5 performs loudness characteristic compensation processing on the synthesized speech by the transversal filter means 21 (step 108), and the bus 8
The data is transferred to the output unit 7 via (step 109).

The DAC 27 of the output section 7 converts the transferred digital voice data into an analog voice signal (step 110). The band limiting filter 28 limits the band of the analog voice (step 111). The output amplifier 29 outputs this.

By the above-mentioned processing, the voice synthesis processing for the voice synthesized by the voice synthesis processing unit 40 and the compensation of the loudness characteristic can be performed according to the application, and the user can listen to the good synthesized voice. .

Third Embodiment The hearing aid of the third embodiment is the first embodiment (FIG. 1).
In addition to the same device configuration as described above, a voice recording / playback processing unit 50 is provided. The sound reproduced by the sound recording / reproducing processing unit 50 is output after being processed by the sound processing processing unit 4 and / or the loudness characteristic compensation processing unit 5.

As a result, the embodiment has the following new features in addition to the features of the first embodiment.

(1) The hearing aid is in the hearing aid mode,
In addition to the hearing check mode, a recording / playback mode for recording and playing back external sound is provided. Therefore, it is possible to record a message to the user himself, record lessons and lectures, and record news on TV and radio.

(2) When playing back the recorded voice,
It is possible to perform voice processing and loudness characteristic compensation processing.

In order to realize the above characteristics, in the third embodiment, a voice recording / playback processing unit 50 is provided in place of the voice synthesis processing unit 40 in the second embodiment (see FIG. 3).

The voice recording / playback processing unit 50 is for recording the voice from the input unit 2 for a long time.

In this embodiment, the voice recording / playback processing unit 50 is used.
As shown in FIG. 5, voice compression means 51 for bit rate compressing the digital voice data and voice demodulation for restoring the bit rate compressed digital voice data to the original voice data. It comprises means 52, a recording memory 54, and a memory interface 53. The memory interface 53 includes a voice compression unit 51 and a voice demodulation unit 52.
And the recording memory 54 exchange data.

[0119]

According to the present invention, the user can measure the frequency characteristic of his own hearing and set the loudness compensation characteristic. Therefore, the user can always maintain the optimum loudness characteristic of the hearing aid by regularly checking his / her own hearing ability, and the usability is improved.

A transformer having a plurality of taps
By using the loudness characteristic compensating means composed of the monkey filter, it is possible to accurately compensate the loudness characteristic.

The processing by the non-linear signal processing means such as the voice processing is performed before the processing by the loudness characteristic compensating means, so that the loudness characteristic compensation can always be provided with high accuracy regardless of the presence or absence of the voice processing. You can

Since the processing by the voice synthesizing means and the external voice recording / reproducing means is performed before the processing by the voice processing means and the above-mentioned loudness characteristic compensating means, the voice synthesizing means and the external voice recording / reproducing means. It is possible to perform voice processing and loudness characteristic compensation on the voice from.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a hearing aid equipped with a loudness characteristic compensation function, which is a first embodiment of the present invention.

FIG. 2 is a diagram showing a loudness characteristic compensation algorithm in the first embodiment.

FIG. 3 is a configuration diagram showing a hearing aid equipped with a loudness characteristic compensation function, which is a second embodiment of the present invention.

FIG. 4 is a diagram showing a flow of processing in a hearing aid mode and a message mode in the second embodiment.

FIG. 5 is a diagram showing a configuration of a hearing aid equipped with a loudness characteristic compensation function, which is a third embodiment of the present invention.

[Explanation of symbols]

1 ... Voice input, 2 ... Input processing unit, 3 ... Memory unit, 4 ... Sound processing unit, 5 ... Loudness characteristic compensation processing unit, 6 ... Hearing check unit, 7 ... Output processing unit, 8 ... Bass, 9 ... Output Terminal, 10 ... Input terminal, 11 ... Band limiting filter, 12 ...
Clamp, 13 ... ADC, 14 ... Memory interface, 15 ... Memory, 16 ... Pitch detecting means, 17 ... Vowel expanding means, 18 ... Consonant detecting means, 19 ... Consonant emphasizing means,
20 ... Combining means, 21 ... Transversal filter means, 22 ... Compensation characteristic calculating means, 23 ... Tap coefficient setting means, 24 ... Response signal taking-in means, 25 ... Test signal generating means, 26 ... Hearing calculating means, 27 ... DAC , 28 ... Band limiting filter, 29 ... Output amplifier, 40 ... Voice synthesis processing section, 50 ... Voice recording / playback processing section, 51 ... Voice compression means,
52 ... Voice demodulation means, 53 ... Memory interface, 5
4 ... Memory for recording

Claims (10)

[Claims] 1. Hearing checking means for measuring a frequency characteristic of a user's hearing, and loudness characteristic compensating means for subjecting a voice signal inputted from the outside to a loudness characteristic compensating process based on a measurement result of the hearing checking means. A loudness compensating apparatus comprising: an output unit that outputs the audio signal that has been subjected to the loudness compensating process to the outside. 2. The loudness compensating means comprises a transversal filter, and a compensation characteristic calculating means for obtaining a loudness compensating frequency characteristic necessary for compensating the frequency characteristic of the hearing of the user obtained by the hearing checking means. And a coefficient setting means for obtaining the tap coefficient value of the transversal filter for realizing the loudness compensation frequency characteristic obtained by the compensation characteristic calculation means with the transversal filter, and setting the tap coefficient value in the transversal filter, The loudness compensating apparatus according to claim 1, wherein the loudness compensating apparatus comprises: 3. A hearing signal checking means is configured to generate a plurality of types of test signals having different frequencies and / or output levels, and a test signal generating means for outputting only one of the test signals at one time. And a response signal capturing means that receives and receives an input of a response signal from a user while the test signal generating means is outputting the test signal, and a frequency of the test signal output at the time of capturing the response signal and its output Based on the level, the user
The hearing aid calculation means for obtaining the frequency characteristic of the hearing ability of the hearing aid is provided, and the output means outputs the test signal output from the test signal generation means and not subjected to the loudness characteristic compensation processing to the outside. The loudness compensating device according to claim 1 or 2, wherein the loudness compensating device outputs. 4. The loudness compensator according to claim 3, wherein said test signal generating means is capable of continuously changing the output level of said test signal. 5. A voice processing unit for performing a non-linear voice processing process on a voice signal input from the outside is provided, and the loudness compensating unit is for the loudness of the signal processed by the voice processing unit. The loudness compensating apparatus according to claim 1, wherein characteristic compensation processing is performed. 6. The non-linear voice processing includes: vowel pitch expansion processing, vowel pitch shortening processing, consonant emphasis processing,
6. The loudness compensator according to claim 5, wherein at least one of the above is included. 7. A voice synthesizing means for synthesizing a voice and at least one of a voice recording / reproducing means for recording a voice signal inputted from the outside, wherein the loudness compensating means comprises the voice synthesizing means and / or
2. The loudness compensator according to claim 1, wherein the loudness characteristic compensating process is performed on the output signal of the audio recording / reproducing means. 8. A voice synthesizing means for synthesizing a voice and at least one of a voice recording / reproducing means for recording a voice signal inputted from the outside, wherein the voice processing means is the voice synthesizing means and the voice synthesizing means. 7. The loudness compensating apparatus according to claim 5, wherein the output signal of the audio recording / reproducing means is subjected to the non-linear audio processing process. 9. Hearing checking means for measuring the frequency characteristic of the user's hearing, and loudness characteristic compensating means for subjecting a voice signal input from the outside to a loudness characteristic compensating process based on the measurement result of the hearing checking means. An output unit that outputs the audio signal that has been subjected to the loudness compensation processing to the outside, and a control unit that has at least a hearing aid mode and a hearing check mode as operation modes, the control unit including: In the hearing aid mode, the loudness characteristic compensating means is caused to perform the loudness characteristic compensating process on the audio signal inputted from the outside, and in the hearing check mode, the frequency characteristic of the hearing ability of the user is measured by the hearing checking means. A hearing aid characterized by: 10. The hearing check means outputs a plurality of types of test signals having different frequencies one by one while gradually increasing the output level thereof, and outputs a response signal from a user during the output of the test signal. Is received and the frequency characteristic of the hearing ability of the user is obtained based on the frequency of the test signal output at the time of receiving the response signal and the output level thereof. Item 9. The hearing aid according to item 9.